Staphylococcus borealis sp. nov., isolated from human skin and blood
Maria Pain1,*, Runa Wolden1, Daniel Jaén- Luchoro2,3, Francisco Salvà-Serra3,4,5,6, Beatriz Piñeiro Iglesias3,4, Roger Karlsson3,4, Claus Klingenberg1,7 and Jorunn Pauline Cavanagh1,7
DOI 10.1099/ijsem.0.004499
Author affiliations: 1Pediatric Infection Group, Department of Clinical Medicine, UiT The Arctic University of Norway, Tromsø, Norway; 2Department of Infectious Diseases, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; 3Centre for Antibiotic Resistance Research (CARe), University of Gothenburg, Gothenburg, Sweden; 4Department of Clinical Microbiology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden; 5Culture Collection University of Gothenburg (CCUG), Sahlgrenska Academy, University of Gothenburg, Sweden;
6Microbiology, Department of Biology, University of the Balearic Islands, Palma de Mallorca, Spain; 7Department of Paediatrics, University Hospital of North Norway, Tromsø, Norway.
*Correspondence: Maria Pain, mcronessen@ gmail. com
Keywords: Coagulase- negative staphylococci; Staphylococcus borealis; Staphylococcus haemolyticus; Staphylococcus; urease; whole- genome sequencing.
Abbreviations: ANI, average nucleotide identity; CFA- FAME, cell fatty acid–fatty acid methyl ester; CoNS, coagulase- negative staphylococci; dDDH, digital DNA–DNA hybridization; GGDC, Genome- to- Genome Distance Calculator; HSP, high- scoring segment pairs; ML, maximum- likelihood; MSA, multi sequence alignment; OGRI, overall genome related index; UBCG, Up- to- date Bacterial Core Gene.
The 16S rRNA gene sequence of Staphylococcus borealis 51-48T is available under the accession number MT586030. The genome sequence data from Staphylococcus borealis 51-48T is available under these accessions: BioSample number SAMN15197055 and assembly accession number GCA_013345165.1. The genome sequence data from Staphylococcus borealis strains 57-14, 57-74, 58-22 and 58-52 are available under BioSample numbers SAMN15197056, SAMN15197057, SAMN15197058 and SAMN15197059, and assembly accession numbers GCA_013345185.1, GCA_013345175.1, GCA_013345195.1 and GCA_013345205.1, respectively.
One supplementary figure and four supplementary tables are available with the online version of this article.
Abstract
When analysing a large cohort of Staphylococcus haemolyticus, using whole- genome sequencing, five human isolates (four from the skin and one from a blood culture) with aberrant phenotypic and genotypic traits were identified. They were phenotypi- cally similar with yellow colonies, nearly identical 16S rRNA gene sequences and initially speciated as S. haemolyticus based on 16S rRNA gene sequence and MALDI- TOF MS. However, compared to S. haemolyticus, these five strains demonstrate: (i) considerable phylogenetic distance with an average nucleotide identity <95 % and inferred DNA–DNA hybridization <70 %; (ii) a pigmented phenotype; (iii) urease production; and (iv) different fatty acid composition. Based on the phenotypic and genotypic results, we conclude that these strains represent a novel species, for which the name Staphylococcus borealis sp. nov. is pro- posed. The novel species belong to the genus Staphylococcus and is coagulase- and oxidase- negative and catalase- positive.
The type strain, 51-48T, is deposited in the Culture Collection University of Gothenburg (CCUG 73747T) and in the Spanish Type Culture Collection (CECT 30011T).
INTRODUCTION
Members of the genus Staphylococcus, currently consisting of 54 species and 22 subspecies with validly published names (based on the List of Prokaryotic Names with Standing in Nomenclature, https:// lpsn. dsmz. de), are most often found on the skin and mucus membranes of mammals and birds [1]. Staphylococci, and particularly the coagulase- positive Staphylococcus aureus, are a major cause of clinical disease in both humans and animals [2–4]. The coagulase- negative staphylococci (CoNS) colonize different niches of the human skin [5] and are part of the commensal human host micro- biota. However, over the last decades some CoNS species such as Staphylococcus epidermidis, Staphylococcus hominis
and Staphylococcus haemolyticus have emerged as important opportunistic pathogens primarily causing disease in patients with foreign body implants or impaired immunity [2].
As part of a previous study analysing a large cohort of S. haemolyticus [6], we detected five bacterial strains with aberrant phenotypic and genotypic traits. All five strains orig- inated from the same geographic location, Tromsø, in North Norway. Four strains were isolated from skin swabs from the groin and armpit of healthy volunteers [5], and one strain was isolated from blood culture in 1997 at the University Hospital of North Norway [7]. The five strains were all initially identi- fied as S. haemolyticus based on 16S rRNA gene sequencing and matrix- assisted laser desorption ionization time- of- flight
mass spectrometry (MALDI- TOF MS) using a Microflex LT instrument (Bruker Daltonics), Flex Control software and the MALDI Biotyper 3.1 software (Bruker Daltonics).
The genome of the blood culture strain was published as S. haemolyticus under the accession number GCA_001224225.1 in 2015 [7].
We decided to perform further analyses of these five strains due to the differences observed in both genotypic and pheno- typic features compared to other S. haemolyticus strains.
Based on results from established phenotypic and genotypic methods for species identification [8, 9], we propose that these five strains belong to a new staphylococcal species hereafter designated Staphylococcus borealis sp. nov.
WHOLE-GENOME SEQUENCING (WGS) AND PHYLOGENETIC ANALYSIS
Genomic DNA from the blood culture strain was isolated according to Chachaty and Saulnier [10] with the addition of RNase A (10 mg ml−1; Qiagen) for Illumina sequencing, and the MasterPure Gram Positive DNA Purification Kit (Lucigen) for isolating genomic DNA for PacBio sequencing.
Genomic DNA from the four commensal isolates was isolated by using the Wizard Genomic DNA kit (Promega). WGS was performed using the Illumina Genome Analyzer II (for the blood culture strain) and Illumina MiSeq (for the four commensal strains), as described previously [6, 7]. Strain 51-48T was additionally sequenced with a PacBio RS II instru- ment (Pacific Biosciences) at the Norwegian Sequencing Centre (NSC), Oslo, Norway. Assembly of Illumina reads was done using Shovill version 0.8.0 (https:// github. com/
tseemann/ shovill). For the PacBio long reads, consensus sequences were generated and assembled with HGAP version 3 (Pacific Biosciences, SMRT Analysis Software version 2.3.0) [11]. The assembly was polished with Pilon version 1.23 [12], using the Illumina sequences (BioSample, SAMEA1035138;
SRA, ERS066311) generated previously by Cavanagh et al.
[7]. Mapping of Illumina sequences and the PacBio assembly were done using BWA- MEM (version 0.715- r1140) [13]. The
resultant draft genome sequences were deposited in GenBank under the BioProject PRJNA638539.
As these strains were initially identified as S. haemolyticus based on both 16S rRNA gene sequence similarity and MALDI- TOF, we performed a core- genome phylogeny anal- ysis on all available S. haemolyticus genomes deposited in the National Center for Biotechnology Information (NCBI) at the time to see whether there were additional isolates related to S. borealis. Two draft genomes isolated from cattle in Canada (SNUC119, assembly accession nos. GCA_003580835.1 and SNUC 1342; assembly accession no. GCA_003042555.1) [14] clustering with the five proposed S. borealis strains were identified. Additionally, a recent publication presented the draft genome of a novel Staphylococcus species isolated from human skin in Denmark (Staphylococcus sp. strain 170 179, accession no. GCA_009735325.1); the authors reported that the closest related genome was the clinical S. borealis strain (51-48T, GCA_001224225.1) [15]. These three additional draft genomes were included in all the comparative genomic analyses. All eight genomes were annotated with Prokka (version 1.13) [16] for downstream analysis.
The genome size range was 2 521 961–2 797 948 bp, with 2288–2529 coding sequences (CDSs). The G+C content of the novel species ranged from 33.54 to 33.80 mol% (Table 1), which is in the range of 33–40 mol% expected for species of the genus Staphylococcus [1]. The G+C content was 0.64–
0.9 mol% higher than that of S. haemolyticus NCTC 11042T. The draft genome of strain 51-48T was 2 797 948 bp long and had 292.2× depth of sequencing coverage. The draft genome had 33.75 mol% G+C content and contained a total of 2529 CDS, 22 rRNA (Among the 22 rRNA, one 5S rRNA and one 16S rRNA were partial sequences) genes (eight copies of each 5S rRNA, seven copies of 16S rRNA and seven copies of 23S rRNA), one tmRNA and 67 tRNAs.
The 16S rRNA gene sequences of S. borealis 51-48T, 57-14, 57-74, 58-22 and 58-52 was determined by Sanger sequencing (forward primer; 5′- TACATGCAAGTCGAGCGAAC-3′ and
Table 1. Overview of genomic information for all eight Staphylococcus borealis strains and the Staphylococcus haemolyticus type strain
Isolate ID Genome size Contigs CDS N50 G+C content
(mol%) Coverage Accession
51-48T(=CCUG 73747T=CECT 30011T) 2 797 948 bp 5 2529 2 689 815 33.75 292.2× GCA_013345165.1 57-14 (=CCUG 73748=CECT 30010) 2 626 230 bp 36 2403 645 817 33.66 337.7× GCA_013345185.1
57-74 (=CCUG 73749) 2 615 713 bp 41 2398 390 616 33.66 463.8× GCA_013345175.1
58-22 (=CCUG 73750) 2 666 192 bp 40 2475 391 465 33.69 319.4× GCA_013345195.1
58-52 (=CCUG 73751) 2 664 706 bp 30 2420 805 534 33.54 387.0× GCA_013345205.1
Staphylococcus sp. 170 179 2 629 435 bp 48 2324 212 499 33.58 334.0× GCA_009735325.1
SNUC119 2 521 961 bp 166 2288 42 538 33.80 50× GCA_003580835.1
SNUC1342 2 522 218 bp 99 2290 119 419 33.80 93× GCA_003042555.1
S. haemolyticus NCTC 11042T 2 569 468 bp 4 2323 2 515 409 32.90 100× GCA_900458595.1
reverse primer 5′- ACCTTCCGATACGGCTACCT-3′). The 16S rRNA sequence of SNUC119, SNUC1342 and 170 179 was retrieved from the genome assemblies. The 16S rRNA gene sequences from the S. borealis strains were analysed using the EzBioCloud online tool 16S- based ID [17]. These results showed that the highest similarities were found to S. haemo- lyticus NCTC 11042T (99.79 % for strain SNUC119, 99.86 % for
strains 51-48T, 57-14, 57-74 and SNUC1342; 99.93 % for 58-22 and 58-52) and Staphylococcus petrasii subsp. jettensis (99.5 % for isolates 51-48T, 57-14, 57-74 and SNUC1342; 99.4 % for 58-22, 58-52 and SNUC119) (Table 2). The full- length 16S rRNA gene was detected in all strains but SNUC119 (96.7 % of full length). For the Danish strain 170 179 the 16S rRNA gene was fragmented with some parts missing, and was omitted
Table 2. Overview of results for species identity of the closest related Staphylococcus species and subspecies, compared to the proposed type strain of Staphylococcus borealis 51-48T
The numbers in brackets are the threshold values for species delineation
Staphylococcal type strains 16S rRNA gene
(98.7 %) ANIb
(<95 %) ANIm
(<95 %) Tetra
(<0.989) dDDH
(<70 %)
S. devriesei NCTC 13828T 99.25 79.82 84.92 0.95612 23.8
S. petrasii subsp. petrasii CCM8418T 99.39 80.10 85.49 0.96349 23.3
S. petrasii subsp croceolyticus CCM8421T 99.39 80.35 85.72 0.96578 23.5
S. petrasii subsp. jettensis SEQ110T 99.51 80.28 85.65 0.96301 23.6
S. petrasii subsp. pragensis NRL/St 12/356T 99.46 80.55 85.74 0.96981 23.6
S. hominis subsp. hominis DSM 20328T 99.25 78.54 85.11 0.95310 22.6
S. hominis subsp. novobiosepticus GTC 1228T 98.83 78.67 85.38 0.95569 23.0
S. haemolyticus NCTC 11042T 99.86 87.40 88.66 0.98571 34.2
SNUC 119 99.93 97.67 98.14 0.99807 82.1
SNUC 1342 100 97.65 98.18 0.99727 82.4
Staphylococcus sp. 170 179 – 99.54 99.80 0.99903 98.0
S. borealis 57-14 100 99.58 99.82 0.99895 97.2
S. borealis 57-74 100 99.56 99.78 0.99879 97.4
S. borealis 58-22 99.93 99.56 99.83 0.99910 98.0
S. borealis 58-52 99.93 99.74 99.81 0.99905 98.1
Table 3. Percent identity between housekeeping genes of S. borealis 51-48T and S. haemolyticus NCTC 11042T and the intraspecies variations for each gene within each species
Intraspecies variation within S. haemolyticus was based on representatives from each phylogenetic group of a diverse collection of S. haemolyticus strains [6]. For the S. borealis strains of human origin (51-48, 57-14, 57–74, 58-22, 58-52 and 170179) all housekeeping genes with the exception of the 16S rRNA gene were identical. Within all housekeeping genes (except 16S rRNA), S. borealis specific conserved bases were observed (specific bases/
SNPs found in all S. borealis strains and in no S. haemolyticus strains)
Housekeeping genes (identity cut- off) tuf (98 %)
[22] gap (96 %)
[21] sodA (97 %)
[23] rpoB (93.6 %)
[24] hsp60 (93 %)
[26] dnaJ (88.8 %)
[25] recA
[27] gyrB
[28] 16S (98.7 %) S. haemolyticus
NCTC 11042T versus S. borealis 51-48T
99.2 % 99.4 % 97.2 % 96.1 % 91.4 % 93.1 % 91.2 % 94.7 % 99.86 %
S. borealis intraspecies variation Conserved SNPs only in S. borealis
100 %
4
99.7–100 %
9
99.8–100 %
12
99.3–100 %
113
99.1–100 %
117
99.4–100 %
61
99–100 %
70
99.5–100 %
83
99.93–100 %
0
S. haemolyticus intraspecies variation
99–100 % 99.3–100 % 98.2–100 % 98.9–100 % 98.3–100 % 97.7–100 % 97.0–100 % 98.7–100 % 99.67–100 %
Fig. 1. Phylogenetic relationship of Staphylococcal type strains and the eight S. borealis strains based on core genes. The maximum- likelihood method was used and bootstrapping was set to 100 replicates, using the RAxML software. Macrococcus caseolyticus was used for rooting the tree.
from the analysis. In total 16 staphylococcal species and subspecies showed 16S rRNA gene identity >98.7 %.
The complete 16S rRNA gene sequences of all staphylococcal type strains were downloaded from the EzBioCloud database [17] and a multi sequence alignment (MSA) was created in mega7 [18] using the software muscle (MUltiple Sequence Comparison by Log- Expectation) [19]. The maximum- likelihood (ML) method was used and bootstrapping was set to 500 replicates, using the RAxML software [20]. Fig.
S1 (available in the online version of this article) shows the reconstructed phylogenetic tree generated from the MSA of the 16S rRNA gene sequence of the staphylococci type strains and accession numbers are listed in Table S1.
We also retrieved 16S rRNA gene sequences from a diverse collection of S. haemolyticus (five representative strains from each of six phylogenetic groups [6]) and compared them to S. borealis. We identified intraspecies variations in 16S rRNA gene among the diverse groups of S. haemolyticus, with some strains having identical 16S rRNA gene sequences to S. borealis, emphasizing that 16S rRNA cannot be used to distinguish between the two species.
As the 16S rRNA gene was unable to discriminate S. haemo- lyticus from S. borealis we investigated whether other single housekeeping genes could distinguish the two species. We analysed the sequence identity between S. haemolyticus NCTC 11042T and S. borealis 51-48T of the following housekeeping
genes: gap [21], tuf [22], sodA [23], rpoB [24], dnaJ [25], hsp60 [26], recA [27] and gyrB [28]. Additionally, we looked at intraspecies variations within the eight S. borealis strains and within a diverse group of 30 S. haemolyticus strains (the same strains as used in 16S rRNA comparison). Even though only the hsp60 gene meets the cut- off criteria for different species, we believe that the genes rpoB, dnaJ, hsp60, recA and gyrB can be used to discriminate between S. haemolyticus and S. borealis as S. borealis contains many unique signature bases (found in all eight S. borealis strains and not identified in any of the tested S. haemolyticus). These results are summarized in Table 3.
We also performed multilocus sequence typing (MLST) using the S. haemolyticus- specific MLST- scheme [29]. All S. borealis strains were non- typeable following this scheme, and we observed variations ranging from 23 to 79 SNPs for each of the seven genes to the closest allele in the MLST data- base, including gaps and insertions, further supporting the identification of a new species.
Genome- based phylogeny plays a central role in taxonomy and phylogeny of bacteria and provides higher resolution than 16S rRNA/single gene phylogeny [30]. WGS compari- sons were performed according to the recommended minimal standards for description of new staphylococcal species [8].
We used the up- to- date bacterial core gene set (UBCG) [30], which produced an alignment based on 92 single- copy core
Fig. 2. SNP- based core- genome phylogenetic tree using the kSNP3 suite, of 169 S. haemolyticus strains and the eight S. borealis strains.
genes extracted from WGS of staphylococcal type strains available in GenBank (accession numbers listed in Table S1).
From the concatenated gene sequences extracted by UBCG, a ML phylogenetic tree was inferred using RAxML (raxmlGUI2 beta) [20] using the GTRGAMMA model and 100 bootstrap replications [31]. The results from the phylogenomic tree confirmed that the eight S. borealis strains belong to a novel CoNS species forming their own well- supported branch
(Fig. 1) most closely related to S. haemolyticus, S. hominis, S. petrasii and Staphylococcus devriesei.
We used the overall genome related index (OGRI) methods to calculate average nucleotide identity (ANI) and tetra- nucleotide analysis using the online tool JSpeciesWS [32].
The digitalDNA–DNA hybridization (dDDH) values were calculated using the Genome- to- Genome Distance Calculator
Table 4. Antibiotic resistance genes
The resistance genes listed in the table can confer resistance to the following antimicrobials: ANT4, aminoglycoside; ble, bleomycin; ermC, erythromycin;
fusC, fusidic acid; mgrA, global regulator (β-lactams and quinolones); qacC, quaternary ammonium compounds; vga(A), streptogramin A lincosamides and/or pleuromutilins identified in the different S. borealis strains, based on the antibiotic databases CARD, megaRes and NCBI. For each resistance gene the percentage identity with the genes identified in the S. borealis strain is presented. For numbers marked in bold the resistance phenotype was also confirmed
ANT4 ble ermC fusC mgrA tetK qacC vga(A)
51-48T 100 % 93 % 98.5 %
57-14 93 % 98.5 %
57-74 93 % 98.5 %
58-22 93 % 98.5 %
58-52 100 % 100 % 93 % 100 % 98.5 %
170 179 100 % 100 % 100 % 93 % 98.5 %
SNUC119 93 % 100 % 98.5 %
SNUC1342 93 % 98.5 %
Fig. 3. Yellow pigmentation of the five S. borealis isolates from this study, in comparison to S. haemolyticus CCUG 7323T shown on P- agar.
Table 5. Biochemical tests, based on APIStaph, API 32 Staph and API Coryne, of the five Staphylococcus borealis isolates and the type strain of Staphylococcus haemolyticus
Staphylococcus borealis Staphylococcus
haemolyticus Culture Collection University of Gothenburg (CCUG) 73 747T 73 748 73 749 73 750 73 751 7323T
Local strain identification number 51-48 57-14 57-74 58-22 58-52 63-42
Test
Glucose GLU + + + + + +
Fructose FRU + + + + + −
Arabinose ARA − − − − − −
Ribose RIB + + + + + +
Mannose MNE − − − − + −
Xylose XYL − − − − − −
Sucrose SAC + + + + + +
Lactose LAC − − − − − −
Turanose TUR − + + − − −
Cellobiose CEL − − − − − −
Maltose MAL + + + + + +
Trehalose TRE + + + + + +
Melibiose MEL − − − − − −
Raffinose RAF − − − − − −
Glycogen GLYG − − − − − −
N- Acetyl- glucosamine NAG + − + + − +
Methyl α- d- glucopiranoside MDG − + + + + −
Mannitol MAN + + + + + −
Xylitol XLT − − − − − −
Nitrate NIT + + + + + +
Acetoin production VP − − + − + +
Novobiocin NOVO − − − − − −
Gelatin GEL − − − − − −
Aesculin ESC + − + − + −
Catalase CAT + + + + + +
Urease URE + + + + + −
N- Acetyl-β-glucosaminidase βNAG − − − − − −
α-Glucosidase αGLU − − − − − −
β-Galactosidase βGAL − − − − − −
β-Glucuronidase βGUR + − − + − +
Alkaline phosphatase PAL + − − + − +
Pyrazinamidase PYZ + + + + + +
Arginine arylamidase ArgA − − − − − −
Continued
(GGDC) version 2.1 [33]. The GGDC results were based on the recommended formula 2 (sum of all identities found in high- scoring segment pairs (HSPs), divided by the overall HSP length), which is independent of genome size. Both the ANI and dDDH values between the S. borealis strains and related staphylococci were much lower than those consid- ered to represent the same species [9]. The results from all OGRI methods confirm that the eight S. borealis strains belong to a novel species, which is related to, but distinctly different from, S. haemolyticus. The OGRI values between the closest- related staphylococcal type strains and the S. borealis strains are summarized in Table 2. The OGRI values between the individual S. borealis strains confirmed that these were different strains and that they belonged to the same species (Tables S2 and S3).
Based on the WGS data we reconstructed a core- genome SNP- based ML tree using the KSNP3 package [34] including 169 S. haemolyticus strains, our five S. borealis strains and the three draft genomes similar to S. borealis found in NCBI. The resultant ML tree clearly demonstrates that the five S. borealis strains and the strains included from the NCBI form a distinct cluster separated from S. haemolyticus (Fig. 2).
Based on the OGRI analysis we clearly see a significant difference in the genomes between S. borealis and the closest- related staphylococcal species. To what extent this represents S. borealis- specific genes or SNP variability within genes of similar functions, the OGRI tools does not answer. Thus, in order to identify genes specific for S. borealis, a genome comparison between the eight S. borealis strains and 169 whole- genome sequences of S. haemolyticus strains [6] was performed. We used the pan- genome tool Roary version 3.11.2 [35] at default settings but changed the parameter for minimum percentage identity for blastp to 70 % in order to identify genes significantly different between the two species.
We then extracted the genes found in all eight S. borealis which were found in 1 % or less of the 169 S. haemolyticus strains. We identified 74 S. borealis specific genes (Table S4), and among these genes we found a urease operon (ureAB- CEFGD). Performing pan- genome analysis with the default minimum percentage for blast at 95 % produced a S. borealis pan- genome of 3267 genes, of which 1480 were S. borealis specific and not shared with any S. haemolyticus strains. The
common core genome between the two species comprised 861 genes, which totals 34.6 % of the average S. borealis gene content.
All eight S. borealis genomes contained capsule- like genes, similar to both the S. haemolyticus described capsule (capA- capG, and capK- capM) and to S. aureus capsule genes (cap5H- cap5J and cap5/8L- cap5/8P). The presence of capsule- like genes was initially identified by performing a local blast of the S. borealis genomes against the virulence factor database (VFDB) [36], and was subsequently manually inspected in all eight strains. The six human- associated strains contained the same capsule- like operon, while the two Canadian animal associated strains had a slightly different type. Whether these are functional genes, and their role in this species remains to be investigated.
Antibiotic resistance genes were identified using the following databases: the Comprehensive Antibiotic Resistance Database (CARD) [37], NCBI AMRFinderPlus [38] and MEGARes [39]. The identified antibiotic resistance genes are summa- rized in Table 4. All eight strains contained a vga(A) gene variant (98.48 % identity). Vga(A) variants confer different levels of resistance to streptogramin A, lincosamides and/
or pleuromutilins [40–42]. All eight strains also harboured mgrA, a global regulator shown to play a role in regulation of virulence factors and contributing to decreased susceptibility to antibiotics like quinolones and β-lactams [43].
PHENOTYPIC TESTS AND METABOLIC PROFILING
The phenotype and metabolic profiling were only performed on the five S. borealis strains identified in our own collection.
Coagulase activity was determined using the Staphaurex Plus Latex Agglutination Test (Thermo Scientific). DNAse activity was tested on DNAse agar with methyl green (Oxoid).
Catalase production was determined by the slide catalase test using hydrogen peroxide, and oxidase activity was determined using the filter paper spot method with 1 % Kovács oxidase reagent. All five S. borealis strains were Gram- stain- positive cocci growing in clusters. They were non- motile on motility agar. All were oxidase-, DNAse-, coagulase- and clumping factor- negative, and catalase- positive. All five strains were
Staphylococcus borealis Staphylococcus
haemolyticus Culture Collection University of Gothenburg (CCUG) 73 747T 73 748 73 749 73 750 73 751 7323T
Local strain identification number 51-48 57-14 57-74 58-22 58-52 63-42
Pyrrolidonyl arylamidase PyrA + + + + + +
Ornithine decarboxilase ODC − − − − − −
Arginine dihydrolase ADH + + + + + +
Table 5. Continued
facultative anaerobic, as determined by using the Brewer thioglycollate medium.
Scanning electron microscopy analyses were performed using a Zeiss Sigma scanning electron microscope (SEM;
Zeiss). Whole cells were fixed with 2.5 % glutaraldehyde and 4 % formaldehyde in PHEM- buffer, before sedimentation onto poly- l- lysin coated coverslips. Samples were further processed according to the protocol of Cocchiaro using the Pelco Biowave (Ted Pella) [44]. Samples were dried in a Leica EM CPD300 and mounted on SEM stubs; gold/palladium was applied with a Polaron Range Sputter Coater. The bacterial cells had a round coccoid shape, appeared in clusters and had a diameter of 650 nm to 1.23 µm.
We tested temperature (4, 15, 30, 37, 42 and 45 °C) and NaCl tolerance (0, 0.5, 1.5, 3, 5, 7.5, 10 and 15 %) according to the protocol by Freney et al. [8], using P- agar plates [1].
The haemolysis assay was performed on blood agar plates (Oxoid). S. haemolyticus CCUG 7323T was included as a reference strain in all tests. All five S. borealis strains were able to grow at 30–42 °C, showed tolerance to NaCl up to 15 % and displayed yellow pigmentation on P- agar plates (Fig. 3). After 24 h of aerobic incubation on horse blood agar at 37 °C, the S. borealis strains formed smooth, circular, raised or slightly convex colonies reaching 3–5 mm in diameter. A clear β-haemolysis (2 mm) was observed in the S. borealis strains and S. haemolyticus CCUG 7323T (1.5 mm) on horse blood agar plates.
Metabolic profiling of the five S. borealis strains and S. haemolyticus CCUG 7323T was performed. The CCUG
STX phenotypic worksheet was followed using the API bacterial identification systems APIStaph, API 32 Staph and API Coryne test (bioMérieux; www. ccug. se/ identification/
worksheets), following the instructions of the manufac- turer. The metabolic profiles are summarized in Table 5.
Biochemically, the five S. borealis strains differed in three tests when compared with S. haemolyticus CCUG 7323T. Briefly, all S. borealis strains were urease- positive, where S. haemolyticus is considered urease- negative. Urease produc- tion is one of the main phenotypic factors differentiating S. haemolyticus from Staphylococcus warneri and Staphy- lococcus saprophyticus [45]. All five S. borealis strains were positive for fermentation of fructose and mannitol, which is a variable trait in S. haemolyticus [46].
Cell fatty acid–fatty acid methyl ester (CFA- FAME) analysis was performed for the five S. borealis strains and the resultant profiles were contrasted with the type strain of S. aureus (type species of the genus Staphylococcus) and two closely related species (S. haemolyticus and S. devriesei) (Table 6). The strains were cultivated on Columbia blood agar base plus 5 % defibri- nated horse blood, at 37 °C, aerobically, overnight (18–24 h).
An approximate biomass of 100 mg from each strain was harvested in early stationary phase to carry out fatty acid methyl ester (FAME) extraction. FAMEs were extracted and washed with alkaline solution after saponification and meth- ylation of the cell biomass, following the protocol detailed by midi [47]. The CFA- FAME profile was determined using an HP 5890 gas chromatograph (Hewlett- Packard) and a stand- ardized protocol similar to the midi Sherlock MIS system [47]
Table 6. Cell fatty acid–fatty acid methyl ester (CFA- FAME) analysis of S. borealis strains (CCUG 73747T, CCUG 73748, CCUG 73749, CCUG 73750, CCUG 73751), showing the ECL (equivalent chain length), name of CFAs and the area per peak (%). Also shown are the corresponding CFA profiles of S. aureus (CCUG 1800T), S. devriesei (CCUG 58238T) and S. haemolyticus (CCUG 7323T). tr denotes ‘trace’ which means a peak has been recorded, but too small to be integrated
ECL 13.618 14.621 14.711 15.626 16.000 16.629 16.722 17.724 17.769 18.000 18.633 18.729 20.000
Peak name of
CFA C14 : 0 iso C15 : 0 iso C15 : 0
antesio C16 : 0 iso C16 : 0 C17 : 0 iso C17 : 0
antesio Summed
feature* C18 : 1 ω9c C18 : 0 C19 : 0
iso C19 : 0 antesio C20 : 0 Species
S. aureus
CCUG 1800T tr 7.3 55.1 1.6 2.0 3.9 19.5 3.3 1.4 2.6 tr 1.3 tr
S. devriesei
CCUG 58238T 0 3.2 52.1 tr 1.3 5.1 25.8 2.6 1.5 2.5 1.0 3.8 tr
S. haemolyticus
CCUG 7323T 1.1 6.9 42,7 1.3 1.8 6.8 18.8 1.2 0 8.5 2.8 5.7 2.5
S. borealis 51-48
CCUG 73747T 1.1 10.0 61.8 tr 1.1 4.9 12.8 tr 1.0 2.3 1.0 1.9 tr
S. borealis 57-14
CCUG 73748 tr 12.0 64.3 tr 1.2 5.6 13.1 0 tr 2.4 tr 1.6 0
S. borealis 57-74
CCUG 73749 1.1 10.9 65.6 tr tr 4.6 12.2 tr tr 2.1 1.3 2.3 tr
S. borealis 58-22
CCUG 73750 1.2 10.8 61.1 1.0 1.2 5.5 13.3 tr tr 2.6 1.1 1.7 tr
S. borealis 58-52
CCUG 73751 1.3 11.7 59.2 1.1 1.7 5.3 13.8 tr tr 3.9 0 1.2 tr
as described previously [48]. CFA- FAME analysis of the five S. borealis strains determined the major CFAs to be long- chain saturated fatty acids, C15 : 0 iso (11 %), C15 : 0 anteiso (63 %) and C17 : 0 anteiso (13 %), while other CFAs observed included C17 : 0 iso (5 %) and C18 : 0 (2.5 %) (Table 5). Preponderance of uneven branched- chain fatty acid pairs with a difference of two carbons, iso/anteiso C15 : 0 and iso/anteiso C17 : 0, were present in all samples; of which the anteiso fatty acids, C15 : 0 anteiso (mean of approximately 63 %) and C17 : 0 anteiso (mean close to 13 %) had higher relative proportions than the iso fatty acids from the same pair, C15 : 0 iso (mean near to 11 %) and C17 : 0 iso (mean of approximately 5 %). Only a few straight chain fatty acids were identified, dominated by stearic acid, C18 : 0 (mean close to 2.5 %). As fatty acids are highly preserved in the bacterial membrane, due to their role in the cellular structure, they are useful markers for bacterial differentiation [49]. Staphylococci exhibit specific CFA- FAME patterns at genus level [50], although characteristic fatty acids at species level [51]. The relative proportion of C15 : 0 anteiso is remarkably higher in S. borealis type strain CCUG 73747T compared to S. haemolyticus CCUG 7323T, showing 61.8 and 47% respectively. The increased amount of C15 : 0 anteiso may regulate membrane fluidity at lower temperatures [52, 53].
The determination of peptidoglycan structure was carried out by the German Collection of Microorganisms and Cell Culture GmbH (DSZM) identification service, as described by Schumann [54]. The total hydrolysate (100 °C, 4 N HCl, 16 h) of the peptidoglycan contained muramic acid (Mur) and the amino acids lysine (Lys), alanine (Ala), serine (Ser), glycine (Gly) and glutamic acid (Glu). Quantification of amino acids by GC/MS of N- heptafluorobutyric amino acid isobutylesters resulted in the following molar ratio: 0.9 Lys : 1.7 Ala : 0.7 Ser : 1.0 Glu : 2.4 Gly : 1.1 Mur. The identity of all amino acids was confirmed by agreement in the gas- chromatographic retention time with those of authentic standards and by char- acteristic mass- spectrometric fragment ions of the derivatives.
After hydrolysis under milder conditions (100 °C, 4 N HCl, 0.75 h), the hydrolysate contained (in addition to the amino acids) the peptides Lys- Ala- Ala (backbone isomer), Lys- Ala (backbone isomer), Mur- Ala, Ala- Glu, Ala- Lys- Gly, Ala- Glu- Lys- Gly, Gly- Gly and Gly- Ser but no Gly- Gly- Gly or Lys- Ser peptide. From these data it was concluded that the strain S. borealis 51-48T (CCUG 737547T) displayed the pepti- doglycan type A3α l- Lys–Gly–Gly–L- Ser–Gly (type A11.3, www. peptidoglycan- types. info).
Antimicrobial resistance testing was performed using the disc diffusion method and the microbroth dilution test according to the eucast guidelines [55]. Briefly, a 0.5 McFarland bacte- rial cell suspension was inoculated on Mueller–Hinton agar plates (Oxoid). Discs or MIC gradient strips were place on the inoculated agar plates and were incubated at 35±1 °C for 16–18 h, and zones of inhibition were measured. The five S. borealis strains were susceptible to the antimicrobial agents cefoxitin (30 µg), ciprofloxacin (5 µg), clindamycin (2 µg), gentamicin (10 µg), linezolid (10 µg), rifampicin, tetracycline (30 µg), trimethoprim- sulfamethoxazole (1.25–23.75 µg), vancomycin (0.015–256 µg), clindamycin (0.125–256 µg) and
lincomycin (0.125–256 µg). 51-48T and 58-52 were resistant to erythromycin (both of which harboured the ermC gene) while 58-52 was also resistant to fusidic acid (harbouring the fusC gene). All strains showed resistance to the pleuromutilin antibiotic tiamulin according to the MIC breakpoints given by Frey et al. [56], which could be conferred by the vga(A) gene.
In conclusion, although the eight S. borealis strains share near- identical 16S rRNA gene sequences to S. haemolyticus NCTC 11042T, and are phylogenetically closely related to S. haemolyticus, there are strong phenotypic and genomic justifications for assigning the strains to a novel species of the genus Staphylococcus, for which the name Staphylococcus borealis sp. nov. is proposed.
These justifications are:
(1) Phylogenetic distance, ANI <95 % and inferred DDH
<70 %.
(2) Genome comparisons.
(3) Pigmented phenotype.
(4) Production of urease.
(5) Different cell fatty acid composition.
DESCRIPTION OF STAPHYLOCOCCUS BOREALIS SP. NOV.
Staphylococcus borealis ( bo. re.a′lis. L. masc. adj. borealis related to the North, boreal).
Colonies are 3–5 mm in diameter, round, smooth and have a yellow tint. The difference in pigmentation between typical S. haemolyticus and S. borealis colonies is particularly evident on different supplemented P- agars (non- supplemented, full fat milk and horse blood) after 48 h at 37 °C. Cells are Gram- stain- positive, coccoid, 650 nm to 1.23 µm in diameter and form clusters. They are facultative anaerobic, coagulase- negative and catalase- positive. They are biochemically nega- tive for fructose and mannitol and positive for production of urease. The major fatty acids are branched fatty acid pairs C15 : 0 iso, C15 : 0 anteiso and C17 : 0 anteiso and C17 : 0 iso, while the straight- chain fatty acid C18 : 0 is present in a much lower amount. The peptidoglycan type is type A11.3.
The five S. borealis strains are deposited to the Culture Collec- tion University of Gothenburg (CCUG), with the following numbers: 51-48T=CCUG 73747T, 57-14=CCUG 73748, 57-74=CCUG 73749, 58-22=CCUG 73750 and 58-52=CCUG 73751. Two strains are deposited to the Spanish Type Culture Collection (CECT) with the following identifiers:
51-48T=CECT 30011T and 57-14=CECT 30010.
Funding information
This study was funded by UiT – The Arctic University of Norway, and funds from the Northern Norway Regional Health Authority (HNF1344-17).
Acknowledgements
The Culture Collection of the University of Gothenburg (CCUG) and its staff are acknowledged for providing reference strains and expert characterization analyses. The CCUG is supported by the Department of Clinical Microbiology, Sahlgrenska University Hospital.
Conflicts of interest
The authors declare that there are no conflicts of interest.
References
1. Götz F, Bannerman T, Schleifer K. The genera Staphylococcus and Macrococcus. In: Dworkin M, Falkow S, Rosenberg E, Schleifer K, Stackebrandt E (editors). The Prokayotes, vol 4: Bacteria: Firmicutes, Cyanobacteria. New York, NY: Springer US; 2006. pp. 5–75.
2. Becker K, Heilmann C, Peters G. Coagulase- negative staphylo- cocci. Clin Microbiol Rev 2014;27:870–926.
3. Tong SYC, Davis JS, Eichenberger E, Holland TL, Fowler VG.
Staphylococcus aureus infections: epidemiology, pathophysiology, clinical manifestations, and management. Clin Microbiol Rev 2015;28:603–661.
4. Fluit AC. Livestock- associated Staphylococcus aureus. Clin Micro- biol Infect 2012;18:735–744.
5. Cavanagh JP, Wolden R, Heise P, Esaiassen E, Klingen- berg C et al. Antimicrobial susceptibility and body site distribution of community isolates of coagulase- negative staphylococci. APMIS 2016;124:973–978.
6. Pain M, Hjerde E, Klingenberg C, Cavanagh JP. Compara- tive genomic analysis of Staphylococcus haemolyticus reveals key to hospital adaptation and pathogenicity. Front Microbiol 2019;10:1–13.
7. Cavanagh JP, Hjerde E, Holden MTG, Kahlke T, Klingenberg C et al.
Whole- genome sequencing reveals clonal expansion of multire- sistant Staphylococcus haemolyticus in European hospitals. J Antimicrob Chemother 2014;69:2920–2927.
8. Freney J, Kloos WE, Hajek V, Webster JA, Bes M et al. Recom- mended minimal standards for description of new staphylococcal species. Int J Syst Evol Microbiol 1999;49:489–502.
9. Chun J, Oren A, Ventosa A, Christensen H, Arahal DR et al.
Proposed minimal standards for the use of genome data for the taxonomy of prokaryotes. Int J Syst Evol Microbiol 2018;68:461–466.
10. Chachaty E, Saulnier P. Isolating chromosomal DNA from. In:
Rapley R (editor). The Nucleic Acid Protocols Handbook. New Jersey:
Humana Press; 2000.
11. Chin CS, Alexander DH, Marks P, Klammer AA, Drake J et al.
Nonhybrid, finished microbial genome assemblies from long- read SMRT sequencing data. Nat Methods 2013;10:563–569.
12. Walker BJ, Abeel T, Shea T, Priest M, Abouelliel A et al. Pilon: an integrated tool for comprehensive microbial variant detection and genome assembly improvement. PLoS One 2014;9:e112963.
13. Li H. Aligning sequence reads, clone sequences and assembly contigs with BWA- MEM. Preprint 2013;00:1–3.
14. Naushad S, Barkema HW, Luby C, Condas LAZ, Nobrega DB et al.
Comprehensive phylogenetic analysis of bovine non- aureus staph- ylococci species based on whole- genome sequencing. Front Micro- biol 2016;7:1990.
15. Stødkilde K, Poehlein A, Brüggemann H. Draft genome sequence of a new staphylococcal species isolated from human skin. Micro- biol Resour Announc 2020;9:e01499–19.
16. Seemann T. Prokka: rapid prokaryotic genome annotation. Bioin- formatics 2014;30:2068–2069.
17. Yoon SH, Ha SM, Kwon S, Lim J, Kim Y et al. Introducing EzBio- Cloud: a taxonomically United database of 16S rRNA gene sequences and whole- genome assemblies. Int J Syst Evol Microbiol 2017;67:1613–1617.
18. Kumar S, Stecher G, Tamura K. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol 2016;33:1870–1874.
19. Edgar RC. Muscle: multiple sequence alignment with high accu- racy and high throughput. Nucleic Acids Res 2004;32:1792–1797.
20. Edler D, Klein J, Antonelli A, Silvestro D. raxmlGUI 2.0 beta: a graphical interface and toolkit for phylogenetic analyses using RAxML. bioRxiv 2019;800912.
21. Yugueros J, Temprano A, Berzal B, Sánchez M, Hernanz C et al.
Glyceraldehyde-3- phosphate dehydrogenase- encoding gene as a useful taxonomic tool for Staphylococcus spp. J Clin Microbiol 2000;38:4351–4355.
22. Martineau F, Picard FJ, Ke D, Paradis S, Roy PH et al. Develop- ment of a PCR assay for identification of staphylococci at genus and species levels. J Clin Microbiol 2001;39:2541–2547.
23. Poyart C, Quesne G, Boumaila C, Trieu- Cuot P. Rapid and accurate species- level identification of coagulase- negative staphylococci by using the sodA gene as a target. J Clin Microbiol 2001;39:4296–4301.
24. Drancourt M, Raoult D. rpoB gene sequence- based identification of Staphylococcus species. J Clin Microbiol 2002;40:1333–1338.
25. Shah MM, Iihara H, Noda M, Song SX, Nhung PH et al. dnaJ gene sequence- based assay for species identification and phylogenetic grouping in the genus Staphylococcus. Int J Syst Evol Microbiol 2007;57:25–30.
26. Goh SH, Potter S, Wood JO, Hemmingsen SM, Reynolds RP et al.
Hsp60 gene sequences as universal targets for microbial species identification. J Clin Microbiol 1996;34:818–823.
27. Landeta G, Reverón I, Carrascosa AV, Rivas Bdelas, Muñoz R. Use of recA gene sequence analysis for the identification of Staphy- lococcus equorum strains predominant on dry- cured hams. Food Microbiol 2011;28:1205–1210.
28. Poirier S, Rué O, Peguilhan R, Coeuret G, Zagorec M et al. Deci- phering intra- species bacterial diversity of meat and seafood spoilage microbiota using gyrB amplicon sequencing: a compara- tive analysis with 16S rDNA V3- V4 amplicon sequencing. PLoS One 2018;13:e0204629–26.
29. Cavanagh JP, Klingenberg C, Hanssen AM, Fredheim EA, Fran- cois P et al. Core genome conservation of Staphylococcus haemo- lyticus limits sequence based population structure analysis. J Microbiol Methods 2012;89:159–166.
30. Na SI, Kim YO, Yoon SH, Ha SM, Baek I et al. UBCG: up- to- date bacterial core gene set and pipeline for phylogenomic tree recon- struction. J Microbiol 2018;56:281–285.
31. Felsenstein J. Confidence Limits on Phylogenies : An Approach Using the Bootstrap. Evolution 1985;39:783–791.
32. Richter M, Rosselló-Móra R, Oliver Glöckner F, Peplies J. JSpe- ciesWS: a web server for prokaryotic species circumscription based on pairwise genome comparison. Bioinformatics 2016;32:929–931.
33. Meier- Kolthoff JP, Auch AF, Klenk HP, Göker M. Genome sequence- based species delimitation with confidence intervals and improved distance functions. BMC Bioinformatics 2013;14:60.
34. Gardner SN, Slezak T, Hall BG. kSNP3.0: SNP detection and phylo- genetic analysis of genomes without genome alignment or refer- ence genome. Bioinformatics 2015;31:2877–2878.
35. Page AJ, Cummins CA, Hunt M, Wong VK, Reuter S et al. Roary:
rapid large- scale prokaryote pan genome analysis. Bioinformatics 2015;31:3691–3693.
36. Liu B, Zheng D, Jin Q, Chen L, Yang J. VFDB 2019: a comparative pathogenomic platform with an interactive web interface. Nucleic Acids Res 2019;47:D687–D692.
37. Jia B, Raphenya AR, Alcock B, Waglechner N, Guo P et al. Card 2017:
expansion and model- centric curation of the comprehensive anti- biotic resistance database. Nucleic Acids Res 2017;45:D566–D573.
38. Feldgarden M, Brover V, Haft DH, Prasad AB, Slotta DJ et al. Vali- dating the AMRFINder tool and resistance gene database by using antimicrobial resistance genotype- phenotype correlations in a collection of isolates. Antimicrob Agents Chemother 2019;63:1–20.
39. Doster E, Lakin SM, Dean CJ, Wolfe C, Young JG et al. MEGARes 2.0: a database for classification of antimicrobial drug, biocide and metal resistance determinants in metagenomic sequence data.
Nucleic Acids Res 2020;48:D561–569.
40. Novotna G, Janata J. A new evolutionary variant of the strep- togramin A resistance protein, Vga(A)LC, from Staphylococcus haemolyticus with shifted substrate specificity towards lincosa- mides. Antimicrob Agents Chemother 2006;50:4070–4076.
41. Tessé S, Trueba F, Berthet N, Hot C, Chesneau O. Resistance genes underlying the LSA phenotype of Staphylococcal isolates from France. Antimicrob Agents Chemother 2013;57:4543–4546.
42. Gentry DR, McCloskey L, Gwynn MN, Rittenhouse SF, Scan- garella N et al. Genetic characterization of Vga ABC proteins conferring reduced susceptibility to pleuromutilins in Staphylo- coccus aureus. Antimicrob Agents Chemother 2008;52:4507–4509.
43. Truong- Bolduc QC, Hooper DC. The transcriptional regulators NorG and MgrA modulate resistance to both quinolones and β-lactams in Staphylococcus aureus. J Bacteriol 2007;189:2996–3005.
44. Cocchiaro JL, Kumar Y, Fischer ER, Hackstadt T, Valdivia RH. Cyto- plasmic lipid droplets are translocated into the lumen of the Chla- mydia trachomatis parasitophorous vacuole. Proc Natl Acad Sci U S A 2008;105:9379–9384.
45. Cunha MdeLRS, Sinzato YK, Silveira LVA. Comparison of methods for the identification of coagulase- negative staphylococci. Mem Inst Oswaldo Cruz 2004;99:855–860.
46. Schleifer KH, Kloos WE. Isolation and characterization of Staphylo- cocci from human skin I. amended descriptions of Staphylococcus epidermidis and Staphylococcus saprophyticus and descriptions of three new species: Staphylococcus cohnii, Staphylococcus haemolyticus, and Staphylococcus xylosus. Int J Syst Bacteriol 1975;25:50–61.
47. Sasser M. Identification of Bacteria by Gas Chromatography of Cellular Fatty Acids, MIDI- Tech Note 101; 1990. pp. 1–6.
48. Zamora L, Fernández- Garayzábal JF, Svensson- Stadler LA, Pala- cios MA, Domínguez L et al. Flavobacterium oncorhynchi sp. nov., a
new species isolated from rainbow trout (Oncorhynchus mykiss).
Syst Appl Microbiol 2012;35:86–91.
49. De Carvalho C, Caramujo M. The various roles of fatty acids. Mole- cules 2018;23:2583.
50. Welch DF. Applications of cellular fatty acid analysis. Clin Microbiol Rev 1991;4:422–438.
51. Kotilainen P, Huovinen P, Eerola E. Application of gas- liquid chro- matographic analysis of cellular fatty acids for species identi- fication and typing of coagulase- negative staphylococci. J Clin Microbiol 1991;29:315–322.
52. Kaneda T. Iso- and anteiso- fatty acids in bacteria: biosyn- thesis, function, and taxonomic significance. Microbiol Rev 1991;55:288–302.
53. Onyango LA, Alreshidi MM. Adaptive metabolism in staphylococci:
survival and persistence in environmental and clinical settings. J Pathog 2018;2018:1–11.
54. Schumann P. Peptidoglycan structure. Methods Microbiol 2011;38:101–129.
55. The European Committee on Antimicrobial Suceptibility Testing.
Breakpoint tables for interpretation of MICs and zone diameters, version 9.0 2019.
56. Frey Y, Rodriguez JP, Thomann A, Schwendener S, Perreten V.
Genetic characterization of antimicrobial resistance in coagulase- negative staphylococci from bovine mastitis milk. J Dairy Sci 2013;96:2247–2257.
Five reasons to publish your next article with a Microbiology Society journal
1. The Microbiology Society is a not-for-profit organization.
2. We offer fast and rigorous peer review – average time to first decision is 4–6 weeks.
3. Our journals have a global readership with subscriptions held in research institutions around the world.
4. 80% of our authors rate our submission process as ‘excellent’ or ‘very good’.
5. Your article will be published on an interactive journal platform with advanced metrics.
Find out more and submit your article at microbiologyresearch.org.
